男男同性性行为人群丙型肝炎病毒感染高风险行为评价工具构建研究

Construction and validation of an assessment tool for screening HCV risky behaviors of men who have sex with men by using decision tree model

目的构建男男同性性行为人群(men who have sex with men,MSM)丙型肝炎病毒(hepatitis C virus,HCV)感染高风险行为评价工具。方法本研究组首先开发HCV感染高风险行为评价工具,2019年12月20日—2020年1月14日利用社交软件平台,通过在线调查收集目标人群相关信息,对该工具进行评估,采用决策树模型进行数据分析。结果HCV感染高风险行为评价工具包含的6个条目并全部纳入树模型,树模型包括5层,27个节点,模型Risk估计量为0.085,模型预测正确率为91.52%,树模型索引图和收益图显示模型拟合良好。重要性评价结果显示,对MSM人群HCV感染风险影响由大至小的条目依次为:HIV结果、毒品使用、性病或相关症状、安全套使用、群交和创伤性操作。结论本研究开发的HCV感染高风险行为评价工具简单、易操作,可用于评价MSM人群的HCV感染高风险行为,为精准行为干预提供科学依据。

Objective To construct and validate an assessment tool for screening HCV risky behaviors of men who have sex with men(MSM)by using decision tree algorithm.Methods The research team first developed a risky behavior assessment tool for HCV infection,which was then validated using data collected from HCV high risk group MSM via Blued,the largest gay social network app through an online survey during December 20,2019 and January 14,2020.The decision tree model was applied for data analysis.Results All6 items included in the HCV risky behavior assessment tool were adopted in the decision tree model which had5 stratum and 27 nodes.The risk value of misclassification probability of tree model was 0.085 and the correct probability of prediction for HCV risky behaviors was 91.52%.Index chart and gain chart showed a decent fitting of the model.The importance evaluation results showed that the risk behaviors of MSM for HCV infection from high to low were HIV status,drug use,sexually transmitted disease and/or related symptoms,condom use,group sex and invasive behavior.Conclusions The assessment tool for HCV risky behaviors developed in this study is simple and easy to manipulate for both healthcare providers and MSM,which can be used for the screening of high risk behaviors of MSM and may provide scientific evidence for HCV intervention.